Ensuring Data Integrity
The iPLATINUM iCleansing process addresses the integrity of existing data, cleansing and deduplication actions and its importation into current corporate business solutions. The process employs a software solution (including hosting if necessary) and consulting services and the adoption of standards to facilitate ongoing data management.
Poor quality data is a function of many factors including:
- Poor previous conversions of legacy systems.
- Manually maintained data typed in an inconsistent format causing problems for an automated data de-duplication process to effectively match duplicates in existing systems.
- Mismatching of data held in more than one application area of a Council system, including things like customer names and addresses, property addresses and property title details.
- Failure to enforce business practices which would prevent the duplication of existing records as new ones are added to the database or certain applications that do not detect existing business details.
Data redundancy will increasingly impact on the quality of information as councils move to centralise names and addresses etc. in one database. This can have significant and potentially detrimental implications for eBusiness engagement within the Council and with the wider community.
A typical iPLATINUM iCleansing process consists of:
- Extraction and loading of various name and address data into an iPLATINUM provided warehouse
- Initial parsing, cleansing and deduplication of data using a combination of solution and site-specific business rules e.g. standard street, suburb postcode combinations, format of council name and specific address
- Presentation of cleansing and deduplication results and opportunities for additional cleansing including review and rule refinement
- Training for council staff to undertake bespoke cleansing and deduplication activities, including tag and match options
- Output of de-duplicated and cleansed data in a format suitable for upload to an existing council solution
- Updating of data using existing deduplication function.